In its most common form, the poly(alkylene oxide) poly(ethylene glycol) (PEG) is a linear polymer terminated at each end with hydroxyl groups:HO—CH2CH2O—(CH2CH2O)n—CH2CH2—OHThis polymer can be represented in a brief form as HO-PEG-OH where it is understood that -PEG- represents the following structural unit:—CH2CH2O—(CH2CH2O)n—CH2CH2—where n typically ranges from approximately 10 to 2000.
PEG is of great utility in a variety of biotechnical and pharmaceutical applications, particularly for drug delivery and modification of drug surfaces to promote nonfouling characteristics.
PEG is not toxic, does not tend to promote an immune response, and is soluble in water and in many organic solvents. The PEG polymer can be covalently attached to insoluble molecules to make the resulting PEG-molecule conjugate soluble. For example, Greenwald, Pendri and Bolikal in J. Org. Chem., 60, 331-336 (1995) recite that the water-insoluble drug taxol, when coupled to PEG, becomes water soluble. Davis et al. in U.S. Pat. No. 4,179,337 recite that proteins coupled to PEG have an enhanced blood circulation lifetime because of a reduced rate of kidney clearance and reduced immunogenicity. The lack of toxicity of the polymer and its rate of clearance from the body are important considerations in pharmaceutical applications. Pharmaceutical applications and many leading references are described in the book by Harris (J. M. Harris, Ed., “Biomedical and Biotechnical Applications of Polyethylene Glycol Chemistry, Plenum, New York, 1992).
PEG is commonly used as methoxy-PEG-OH, or mPEG in brief, in which one terminus is the relatively inert methoxy group, while the other terminus is a hydroxyl group that is subject to ready chemical modificationCH3O—(CH2CH2O)n—CH2CH2—OH mPEG
PEG is also commonly used in branched forms that can be prepared by addition of ethylene oxide to various polyols, including glycerol, pentaerythritol and sorbitol. For example, the four-armed branched PEG prepared from pentaerythritol is shown below:C(CH2OH)n+nC2H4O→C[CH2O—(CH2CH2O)n—CH2CH2—OH]4 
The branched PEGs can be represented in a general form as R(-PEG-OH)n in which R represents the central core molecule, which can include, e.g., glycerol or pentaerythritol, and n represents the number of arms.
Often it is necessary to use an “activated derivative” of PEG to couple PEG to a molecule. The hydroxyl group located at the PEG terminus, or other group subject to ready chemical modification, is activated by modifying or replacing the group with a functional group suitable for reacting with a group on another molecule, including, e.g., proteins, surfaces, enzymes, and others. For example, the succinimidyl “active ester” of carboxymethylated PEG forms covalent bonds with amino groups on proteins as described by K. Iwasaki and Y. Iwashita in U.S. Pat. No. 4,670,417. The synthesis described in U.S. Pat. No. 4,670,417 is illustrated below with the active ester reacting with amino groups of a protein in which the succinimidyl group is represented as NHS and the protein is represented as PRO-NH2:PEG-O—CH2—CO2—NHS+PRO-NH2→PEG-O—CH2—CO2—NH-PROSuccinimidyl “active esters”, such as PEG-OCH2—CO2—NHS, are commonly used forms of activated carboxylic acid PEGs, and they are prepared by reacting carboxylic acid PEGs with N-hydroxysuccinimide.
PEG hydrogels, which are water-swollen gels, have been used for wound covering and drug delivery. PEG hydrogels are prepared by incorporating the soluble, hydrophilic polymer into a chemically crosslinked network or matrix so that addition of water produces an insoluble, swollen gel. Substances useful as drugs typically are not covalently attached to the PEG hydrogel for in vivo delivery. Instead, the substances are trapped within the crosslinked matrix and pass through the interstices in the matrix. The insoluble matrix can remain in the body indefinitely, and control of the release of the drug typically can be somewhat imprecise.
One approach to preparation of these hydrogels is described by Embrey and Grant in U.S. Pat. No. 4,894,238. The ends of the linear polymer are connected by various strong, nondegradable chemical linkages. For example, linear PEG is incorporated into a crosslinked network by reacting with a triol and a diisocyanate to form hydrolytically stable urethane linkages that are nondegradable in water.
A related approach for preparation of PEG hydrogels has been described by Gayet and Fortier in J. Controlled Release, 38, 177-184 (1996) in which linear PEG was activated as the p-nitrophenylcarbonate and crosslinked by reaction with a protein, bovine serum albumin. The linkages formed are hydrolytically stable urethane groups and the hydrogels are nondegradable in water.
In another approach, described by N. S. Chu in U.S. Pat. No. 3,963,805, nondegradable PEG networks have been prepared by random entanglement of PEG chains with other polymers formed by use of free radical initiators mixed with multifunctional monomers. P. A. King described nondegradable PEG hydrogels in U.S. Pat. No. 3,149,006 that have been prepared by radiation-induced crosslinking of high molecular weight PEG.
Nagaoka et al. described in U.S. Pat. No. 4,424,311 preparing PEG hydrogels by copolymerization of PEG methacrylate with other comonomers such as methyl methacrylate. Vinyl polymerization produces a polyethylene backbone with PEG attached. The methyl methacrylate comonomer is added to give the gel additional physical strength.
Sawhney et al. described, in Macromolecules, 26, 581 (1993) and U.S. Pat. No. 5,626,863, the preparation of block copolymers of polyglycolide or polylactide and PEG that are terminated with acrylate groups:CH2═CH—CO—(O—CHR—CO)n—O-PEG-O—(CO—CHR—O)n—OC—CH═CH2 where R is CH3— or H.
In the above formula, the glycolide blocks are the —OCH2—CO— units; addition of a methyl group to the methylene group gives rise to a lactide block; n can be multiples of 2. Vinyl polymerization of the acrylate groups produces an insoluble, crosslinked gel with a polyethylene backbone. The polylactide or polyglycolide segments of the polymer backbone shown above, which are ester groups, are susceptible to slow hydrolytic breakdown, with the result that the crosslinked gel undergoes slow degradation and dissolution. While this approach provides for degradable hydrogels, the structure provides no possibility of covalently attaching proteins or other drugs to the hydrogel for controlled release. Applications of these hydrogels in drug delivery are thus restricted to release of proteins or other drugs physically entrapped within the hydrogel, thus reducing the potential for advantageous manipulation of release kinetics.
Hubbell, Pathak, Sawhney, Desai, and Hill (U.S. Pat. No. 5,410,016, 1995) polymerized:Protein-NH-PEG-O2C—CH═CH2with long wavelength uv radiation to obtain a PEG acrylate polymer with a protein linked to it. The link between the PEG and the protein was not degradable, so the protein could only be hydrolytically released with PEG attached. Since the acrylate polymer is not hydrolytically degradable, the release of the PEG protein derivative is not controllable.
Yang, Mesiano, Venkatasubramanian, Gross, Harris and Russell in J. Am. Chem. Soc. 117, 4843-4850, (1995) described heterobifunctional poly(ethylene glycols) having an acrylate group on one terminus and an activated carboxylic acid on the second terminus. They demonstrated the attachment of this PEG derivative to a protein and incorporation of the resulting PEG protein derivative into an acrylate polymer. However, the PEG backbone there is not degradable and the protein was thus, in effect, permanently bound to the acrylate polymer.